make sure all identifiers are valid (#14)

go.mod

@@ -5,6 +5,8 @@ go 1.12
 require (
 	github.com/dave/jennifer v1.3.0
 	github.com/inconshreveable/mousetrap v1.0.0 // indirect
+	github.com/kylelemons/godebug v1.1.0 // indirect
 	github.com/spf13/cobra v0.0.3
 	github.com/spf13/pflag v1.0.3 // indirect
+	golang.org/x/text v0.3.7
 )

go.sum

@@ -2,7 +2,12 @@ github.com/dave/jennifer v1.3.0 h1:p3tl41zjjCZTNBytMwrUuiAnherNUZktlhPTKoF/sEk=
 github.com/dave/jennifer v1.3.0/go.mod h1:fIb+770HOpJ2fmN9EPPKOqm1vMGhB+TwXKMZhrIygKg=
 github.com/inconshreveable/mousetrap v1.0.0 h1:Z8tu5sraLXCXIcARxBp/8cbvlwVa7Z1NHg9XEKhtSvM=
 github.com/inconshreveable/mousetrap v1.0.0/go.mod h1:PxqpIevigyE2G7u3NXJIT2ANytuPF1OarO4DADm73n8=
+github.com/kylelemons/godebug v1.1.0 h1:RPNrshWIDI6G2gRW9EHilWtl7Z6Sb1BR0xunSBf0SNc=
+github.com/kylelemons/godebug v1.1.0/go.mod h1:9/0rRGxNHcop5bhtWyNeEfOS8JIWk580+fNqagV/RAw=
 github.com/spf13/cobra v0.0.3 h1:ZlrZ4XsMRm04Fr5pSFxBgfND2EBVa1nLpiy1stUsX/8=
 github.com/spf13/cobra v0.0.3/go.mod h1:1l0Ry5zgKvJasoi3XT1TypsSe7PqH0Sj9dhYf7v3XqQ=
 github.com/spf13/pflag v1.0.3 h1:zPAT6CGy6wXeQ7NtTnaTerfKOsV6V6F8agHXFiazDkg=
 github.com/spf13/pflag v1.0.3/go.mod h1:DYY7MBk1bdzusC3SYhjObp+wFpr4gzcvqqNjLnInEg4=
+golang.org/x/text v0.3.7 h1:olpwvP2KacW1ZWvsR7uQhoyTYvKAupfQrRGBFM352Gk=
+golang.org/x/text v0.3.7/go.mod h1:u+2+/6zg+i71rQMx5EYifcz6MCKuco9NR6JIITiCfzQ=
+golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=

pkg/generator/generator.go

@@ -5,10 +5,12 @@ import (
 	"errors"
 	"fmt"
 	"sort"
-	"strconv"
 	"strings"
+	"unicode"
 
 	"github.com/marhaupe/json2struct/pkg/parse"
+	"golang.org/x/text/cases"
+	"golang.org/x/text/language"
 
 	"github.com/dave/jennifer/jen"
 )
@@ -247,20 +249,58 @@ func mergeObjects(children []*parse.ObjectNode) *parse.ObjectNode {
 	}
 }
 
+var caser = cases.Title(language.English)
+
 func makeVarname(varname string) *jen.Statement {
-	upperCaseVarname := strings.Title(strings.ToLower(varname))
+	upperCaseVarname := caser.String(strings.ToLower(varname))
+
+	// Maybe we can find a shortcut? This is getting a bit computation heavy
+	if isValid, validIdentifier := identifierIsValid(varname); !isValid {
+		upperCaseVarname = validIdentifier
+		fmt.Println("Invalid identifier found. We cleaned that up for you, but you might want to double check if you're happy with our naming: ", validIdentifier)
+	}
 
-	// Numbers are not a valid identifier.
-	if isNumber(varname) {
-		upperCaseVarname = "Number" + upperCaseVarname
+	if identifierIsPredeclared(varname) {
+		upperCaseVarname = "_" + upperCaseVarname
+		fmt.Println("Predeclared identifier found. We cleaned that up for you, but you might want to double check if you're happy with our naming: ", upperCaseVarname)
 	}
 
 	return jen.Id(upperCaseVarname)
 }
 
-func isNumber(varname string) bool {
-	_, err := strconv.ParseFloat(varname, 64)
-	return err == nil
+// Valid identifiers are specified here: https://go.dev/ref/spec#Identifiers.
+// Summarized we can say:
+// identifier 	  = letter { letter | unicode_digit } .
+// letter         = unicode_letter | "_" .
+// unicode_letter = /* a Unicode code point classified as "Letter" */ .
+// unicode_digit  = /* a Unicode code point classified as "Number, decimal digit" */ .
+func identifierIsValid(identifier string) (bool, string) {
+	characters := []rune(identifier)
+	if !isLetter(characters[0]) {
+		characters[0] = '_'
+	}
+	for i, char := range characters[1:] {
+		if !(isLetter(char) || unicode.IsDigit(char)) {
+			characters[i] = '_'
+		}
+	}
+	validIdentifier := string(characters)
+	return validIdentifier == identifier, validIdentifier
+}
+
+func isLetter(letter rune) bool {
+	return unicode.IsLetter(letter) || letter == '_'
+}
+
+var predeclaredIdentifiers = []string{"any", "bool", "byte", "comparable", "complex64", "complex128", "error", "float32", "float64", "int", "int8", "int16", "int32", "int64", "rune", "string", "uint", "uint8", "uint16", "uint32", "uint64", "uintptr", "true", "false", "iota", "nil", "append", "cap", "close", "complex", "copy", "delete", "imag", "len", "make", "new", "panic", "print", "println", "real", "recover"}
+
+func identifierIsPredeclared(identifier string) bool {
+	for _, predeclared := range predeclaredIdentifiers {
+		if identifier == predeclared {
+			return true
+		}
+	}
+	return false
 }
 
 // addJSONTag adds the json-tag, e.g. `json:"title"`. This has to match the original varname from the json file

pkg/generator/generator_test.go

@@ -6,47 +6,65 @@ import (
 	"path"
 	"strings"
 	"testing"
+
+	"github.com/kylelemons/godebug/diff"
 )
 
-func Test_isNumber(t *testing.T) {
+func Test_identifierIsValid(t *testing.T) {
 	type args struct {
 		varname string
 	}
 	tests := []struct {
-		name string
-		args args
-		want bool
+		name                  string
+		args                  args
+		wantIsValid           bool
+		wantCleanedIdentifier string
 	}{
 		{
-			name: "floating",
-			args: args{"1.1"},
-			want: true,
+			name:        "floating",
+			args:        args{"1.1"},
+			wantIsValid: false,
+		},
+		{
+			name:        "negative floating",
+			args:        args{"-1.1"},
+			wantIsValid: false,
+		},
+		{
+			name:        "int",
+			args:        args{"1"},
+			wantIsValid: false,
+		},
+		{
+			name:        "negative int",
+			args:        args{"-1"},
+			wantIsValid: false,
 		},
 		{
-			name: "negative floating",
-			args: args{"-1.1"},
-			want: true,
+			name:        "$",
+			args:        args{"$test"},
+			wantIsValid: false,
 		},
 		{
-			name: "int",
-			args: args{"1"},
-			want: true,
+			name:        "only letters",
+			args:        args{"xyz"},
+			wantIsValid: true,
 		},
 		{
-			name: "negative int",
-			args: args{"-1"},
-			want: true,
+			name:        "underscore",
+			args:        args{"_test"},
+			wantIsValid: true,
 		},
 		{
-			name: "not number",
-			args: args{"xyz"},
-			want: false,
+			name:        "invalid character in the middle",
+			args:        args{"_$test"},
+			wantIsValid: true,
 		},
 	}
 	for _, tt := range tests {
 		t.Run(tt.name, func(t *testing.T) {
-			if got := isNumber(tt.args.varname); got != tt.want {
-				t.Errorf("isNumber() = %v, want %v", got, tt.want)
+			if got, _ := identifierIsValid(tt.args.varname); got != tt.wantIsValid {
+				t.Errorf("identifierIsValid() = %v, want %v", got, tt.wantIsValid)
 			}
 		})
 	}
@@ -77,8 +95,7 @@ func TestFiles(t *testing.T) {
 		actual = string(formatActualBytes)
 
 		if actual != expected {
-			t.Errorf("Test failed. Filename: %v\nActual: %v\nActualLen: %v\nExpected: %v\nExpectedLen: %v\n",
-				filename, actual, len(actual), expected, len(expected))
+			t.Errorf("Test failed. \nFilename: %v \nDiff: \n\n%v", filename, diff.Diff(actual, expected))
 		}
 	}
 }

pkg/generator/testdata/11_expected

@@ -1,5 +1,5 @@
-package generated 
+package generated
 
-type JSONToStruct struct{
-  Number1234 string `json:"1234"`
-}
+type JSONToStruct struct {
+	_234 string `json:"1234"`
+}

pkg/generator/testdata/with_ invalid_characters

@@ -0,0 +1,26 @@
+{
+  "contentType": "application/vnd.microsoft.card.adaptive",
+  "content": {
+    "type": "AdaptiveCard",
+    "body": [
+      {
+        "type": "TextBlock",
+        "text": "Hi <at>John Doe</at>"
+      }
+    ],
+    "$schema": "https://adaptivecards.io/schemas/adaptive-card.json",
+    "version": "1.0",
+    "msteams": {
+      "entities": [
+        {
+          "type": "mention",
+          "text": "<at>John Doe</at>",
+          "mentioned": {
+            "id": "29:123124124124",
+            "name": "John Doe"
+          }
+        }
+      ]
+    }
+  }
+}

pkg/generator/testdata/with_ invalid_characters_expected

@@ -0,0 +1,24 @@
+package generated
+
+type JSONToStruct struct {
+	Content struct {
+		_schema string `json:"$schema"`
+		Body    []struct {
+			Text string `json:"text"`
+			Type string `json:"type"`
+		} `json:"body"`
+		Msteams struct {
+			Entities []struct {
+				Mentioned struct {
+					Id   string `json:"id"`
+					Name string `json:"name"`
+				} `json:"mentioned"`
+				Text string `json:"text"`
+				Type string `json:"type"`
+			} `json:"entities"`
+		} `json:"msteams"`
+		Type    string `json:"type"`
+		Version string `json:"version"`
+	} `json:"content"`
+	Contenttype string `json:"contentType"`
+}